Microsoft Security Intelligence

Report
Volume 10

An in-depth perspective on
software vulnerabilities and exploits,
malicious code threats, and
potentially unwanted software
in 2010. With new data covering
July through December
Microsoft Security Intelligence Report
This document is for informational purposes only. MICROSOFT MAKES NO
WARRANTIES, EXPRESS, IMPLIED, OR STATUTORY, AS TO THE
INFORMATION IN THIS DOCUMENT.

This document is provided “as-is.” Information and views expressed in this

document, including URL and other Internet website references, may change
without notice. You bear the risk of using it.

Copyright © 2011 Microsoft Corporation. All rights reserved.

The names of actual companies and products mentioned herein may be the
trademarks of their respective owners.

1
 Authors
Doug Cavit Michelle Meyer Javier Salido Jossie Tirado Arroyo
Microsoft Trustworthy Microsoft Trustworthy Microsoft Trustworthy Microsoft IT Information
Computing Computing Computing Security and Risk
Management
Joe Faulhaber Daryl Pecelj Christian Seifert
Microsoft Malware Protection Microsoft IT Information Bing Scott Wu
Center Security and Risk Microsoft Malware Protection
Management Frank Simorjay Center
Vinny Gullotto Microsoft Trustworthy
Microsoft Malware Protection Anthony Penta Computing Jeff Williams
Center Microsoft Windows Safety Microsoft Malware Protection
Platform Holly Stewart Center
Jeff Jones Microsoft Malware Protection
Microsoft Trustworthy Tim Rains Center Terry Zink
Computing Microsoft Trustworthy Microsoft Forefront Online
Computing Matt Thomlinson Protection for Exchange
Jimmy Kuo Microsoft Security Response
Microsoft Malware Protection Center
Center

Contributors
Lawren Ahuna Yuhui Huang Don Nguyen Marc Seinfeld
Microsoft IT Information Microsoft Malware Protection Microsoft IT Information Microsoft Malware Protection
Security and Risk Center Security and Risk Center
Management Management
CSS Japan Security Jasmine Sesso
Eva Chow Response Team Price Oden Microsoft Malware Protection
Microsoft IT Information Microsoft Japan Microsoft IT Information Center
Security and Risk Security and Risk
Management John Lambert Management Norie Tamura (GOMI)
Microsoft Security CSS Japan Security Response
Enrique Gonzalez Engineering Center Kathy Phillips Team
Microsoft Malware Protection Microsoft Legal and
Center Eric Leonard Corporate Affairs Gilou Tenebro
Microsoft IT Information Microsoft Malware Protection
Cristin Goodwin Security and Risk Hilda Larina Ragragio Center
Microsoft Legal and Management Microsoft Malware Protection
Corporate Affairs Center
Laura Lemire
Satomi Hayakawa Microsoft Legal and Tareq Saade
CSS Japan Security Response Corporate Affairs Microsoft Malware Protection
Team Center
Ken Malcolmson
Microsoft Trustworthy Richard Saunders
Computing Microsoft Trustworthy
Computing
Charles McColgan
Microsoft ISD

2
Table of Contents

About This Report ..............................................................................................5

Scope .............................................................................................................5
Reporting Period ............................................................................................5
Conventions ...................................................................................................5
Key Findings Summary ......................................................................................6
Vulnerability Disclosures ................................................................................6
Exploits ..........................................................................................................6
Malware and Potentially Unwanted Software ..................................................7
Operating System Infection Rates....................................................................7
Threat Families ...............................................................................................7
Home and Enterprise Threats .........................................................................8
Email Threats .................................................................................................8
Spam Types ....................................................................................................9
Malicious Websites .........................................................................................9
Trustworthy Computing: Security Engineering at Microsoft .............................. 11
Vulnerabilities .................................................................................................. 12
Vulnerability Severity ................................................................................... 12
Vulnerability Complexity .............................................................................. 14
Operating System, Browser, and Application Vulnerabilities ......................... 15
Vulnerability Disclosures .............................................................................. 16
Exploits ............................................................................................................ 18
HTML and JScript/JavaScript Exploits ........................................................... 20
Document Exploits ....................................................................................... 21
Operating System Exploits ............................................................................ 22

3
 Security Breach Trends ................................................................................. 24
Malware and Potentially Unwanted Software .................................................... 27
Global Infection Rates ................................................................................... 27
Operating System Infection Rates.................................................................. 33
Threat Categories.......................................................................................... 36
Threat Categories by Location ................................................................... 37
Threat Families ............................................................................................. 39
Rogue Security Software ............................................................................... 41
Home and Enterprise Threats ....................................................................... 45
Email Threats ................................................................................................... 49
Spam Messages Blocked ................................................................................ 49
Spam Types .................................................................................................. 51
Malicious Websites ........................................................................................... 55
Phishing Sites ............................................................................................... 56
Target Institutions..................................................................................... 57
Global Distribution of Phishing Sites ......................................................... 59
Malware Hosting Sites .................................................................................. 61
Malware Categories ................................................................................... 62
Global Distribution of Malware Hosting Sites ............................................ 65
Drive-By Download Sites .............................................................................. 66
Appendix A: Threat Naming Conventions ........................................................ 69
Appendix B: Data Sources ................................................................................ 71
Microsoft Products and Services ................................................................ 71
Appendix C: Worldwide Infection Rates ........................................................... 73
Glossary ........................................................................................................... 78
Threat Families Referenced in This Report ........................................................ 83

4
About This Report

Scope
The Microsoft® Security Intelligence Report (SIR) focuses on software
vulnerabilities, software vulnerability exploits, malicious and potentially
unwanted software, and security breaches. Past reports and related resources are
available for download at www.microsoft.com/sir. We hope that readers find the
data, insights, and guidance provided in this report useful in helping them
protect their organizations, software, and users.

Reporting Period
In this volume of the Microsoft Security Intelligence Report, statistics about malware
families and infections are reported on a quarterly basis and other statistics
continue to be reported on a half-yearly basis, with a focus on 2010.

Throughout the report, half-yearly and quarterly time periods are referenced
using the nHyy or nQyy formats, respectively, where yy indicates the calendar
year and n indicates the half or quarter. For example, 1H10 represents the first
half of 2010 (January 1 through June 30), and 2Q10 represents the second
quarter of 2010 (April 1 through June 30). To avoid confusion, please pay
attention to the reporting period or periods being referenced when considering
the statistics in this report.

Conventions
This report uses the Microsoft Malware Protection Center (MMPC) naming
standard for families and variants of malware and potentially unwanted software.
For information about this standard, see “Microsoft Malware Protection Center
Naming Standard” on the MMPC website.

5
 Key Findings Summary

Volume 10 of the Microsoft® Security Intelligence Report (SIRv10) provides in-

depth perspectives on software vulnerabilities, software vulnerability exploits,
malicious and potentially unwanted software, and security breaches in both
Microsoft and third party software. Microsoft developed these perspectives based
on detailed trend analysis over the past several years, with a focus on 2010.

This document summarizes the key findings of the report. The full SIRv10 also
includes deep analysis of trends found in 117 countries/regions around the world
and offers ways to manage risks to your organization, software, and people.

The full SIRv10, as well as previous volumes of the report and related videos, can
be downloaded from www.microsoft.com/sir.

Vulnerability Disclosures
 Vulnerabilities in applications versus operating systems or web browsers
continued to account for a large majority of all vulnerabilities in 2010,
although the total number of application vulnerabilities declined 22.2
percent from 2009.

 Industry vulnerability disclosure trends continue an overall trend of

moderate declines since 2006. This trend is likely because of better
development practices and quality control throughout the industry,
which result in more secure software and fewer vulnerabilities.

 Vulnerability disclosures for Microsoft products increased slightly in

2010 but have generally remained stable over the past several periods.

Exploits
 The exploitation of Java vulnerabilities sharply increased in the third
quarter of 2010 and surpassed every other exploitation category that the

6
 MMPC tracks, including generic HTML/scripting exploits, operating
system exploits, and document exploits.
 Exploits that use HTML and JavaScript steadily increased throughout the
year and continue to represent a large portion of exploits. The most
prevalent type of attack in this category involved malicious IFrames.
 The number of Adobe Acrobat and Adobe Reader exploits dropped by
more than half after the first quarter and remained near this reduced
level throughout the remainder of the year.

Malware and Potentially Unwanted Software

 Except where specified, the information in this section was compiled
from telemetry data that was generated from more than 600 million
computers worldwide and some of the busiest online services on the
Internet.

Operating System Infection Rates

 As in previous periods, infection rates for more recently released
Microsoft operating systems and service packs are consistently lower
than older ones, for both client and server platforms. Windows 7 and
Windows Server 2008 R2, the most recently released Windows client
and server versions, respectively, have the lowest infection rates.

 Infection rates for the 64-bit versions of Windows Vista® and Windows
7 are lower than for the corresponding 32-bit versions of those operating
systems. One reason may be that 64-bit versions of Windows still appeal
to a more technically savvy audience than their 32-bit counterparts,
despite increasing sales of 64-bit Windows versions among the general
computing population. Kernel Patch Protection (KPP), a feature of 64-bit
versions of Windows that protects the kernel from unauthorized
modification, may also contribute to the difference by preventing certain
types of malware from operating.

Threat Families
 JS/Pornpop, the most commonly detected family in 4Q10, is a detection
for specially crafted JavaScript-enabled objects that attempt to display

7
 pop-under advertisements in users’ web browsers, usually with adult
content.

 Detections and removals of Win32/Autorun, a generic detection for

worms that spread between mounted volumes using the Autorun feature
of Windows, increased significantly in 4Q10, although Autorun dropped
to second place because of the spread of Pornpop.

 Win32/Taterf, the most prevalent threat in 2Q10, dropped to third by

4Q10. Taterf belongs to a category of threats that are designed to steal
passwords for popular online computer games and transmit them to the
attackers. See “Online Gaming-Related Families” on page 62 of Microsoft
Security Intelligence Report, Volume 5 (January through June 2008) for more
information about these threats.

Home and Enterprise Threats

 Seven malware families are common to home and enterprise network
environments, although they are ordered differently and in different
proportions. The worm family Win32/Conficker, which uses several
methods of propagation that work more effectively within a typical
enterprise network environment than they do over the public Internet,
leads the domain-joined list by a significant margin, but ranks ninth on
the non-domain list.

 On non-domain computers, JS/Pornpop was the most commonly

detected family in 4Q10 and the fourth most commonly detected family
in 2010 overall. By contrast, this family was detected much less often on
domain-joined computers. Pornpop is often found on websites that host
illegal or illicit content, which users in domain environments are often
restricted from accessing by organizational policy or blocking software.

Email Threats
 After increasing gradually and then reaching a plateau through the first
eight months of 2010, the number of spam messages received and
blocked by Microsoft Forefront® Online Protection for Exchange
(FOPE) dropped abruptly in September, and again in December. These

8
 drops can be correlated with events involving two of the world’s most
significant spam-sending botnets:

o During the last week of August 2010, researchers affiliated with the
security firm LastLine spearheaded a coordinated takedown of
command-and-control (C&C) servers associated with the
Win32/Cutwail spambot. In the days following the takedown, FOPE
recorded a significant drop in the average daily volume of messages
blocked.

o On or about December 25, 2010, spam researchers around the world

recorded an almost complete cessation of spam originating from the
large Rustock botnet, with some spam trackers reporting a drop in
the global spam rate as high as 50 percent or more. During the final
week of December, the number of messages blocked by FOPE was
almost 30 percent less than in the prior week, compared to a drop of
less than two percent between the final two weeks of 2009. The
Rustock botnet subsequently began sending spam again in mid-
January, and the number of messages blocked by FOPE has risen
accordingly. The reasons for this hiatus are still being investigated.

Spam Types
 Advertisements for nonsexual pharmaceutical products accounted for
32.4 percent of the spam messages blocked by FOPE content filters in
2010.

 Together with nonpharmaceutical product ads (18.3 percent of the total)

and advertisements for sexual performance products (3.3 percent),
product advertisements accounted for 54.0 percent of spam in 2010,
which is down from 69.2 percent a year ago.

Malicious Websites
 In the first half of 2010, phishers showed signs of targeting online
gaming sites with increasing frequency, although this push appeared to
have dwindled as social networks came under increased attack.
Impressions that targeted gaming sites reached a high of 16.7 percent of

9
 all impressions in June before dropping to a more typical 2.1 percent in
December.

 Phishing sites that target social networks routinely receive the highest
number of impressions per active phishing site. The percentage of active
phishing sites that targeted social networks increased during the final
months of the year, but still only accounted for 4.2 percent of active sites
in December, despite receiving 84.5 percent of impressions that month.
Nevertheless, the number of active sites targeting gaming sites remained
relatively high during the second half of the year, which suggests that
more campaigns may be coming.

10
Trustworthy Computing:
Security Engineering at
Microsoft

Amid the increasing complexity of today’s computing threat landscape and the
growing sophistication of criminal attacks, enterprises and governments are more
focused than ever on protecting their computing environments so that they and
their constituents can feel safer online. With more than a billion systems using its
products and services worldwide, Microsoft collaborates with partners, industry,
and governments to help create a safer, more trusted Internet.

Trustworthy Computing (TwC), formed in 2002, is Microsoft’s commitment to

creating and delivering secure, private, and reliable computing experiences based
on sound business practices. The intelligence provided in this report comes from
Trustworthy Computing security centers that deliver in-depth threat intelligence,
threat response, and security science, as well as information from product groups
across Microsoft. The report is designed to give our customers, partners, and the
industry a better understanding of the threat landscape so that they will be in a
better position to protect themselves and their assets from criminal activity.

11
 Vulnerabilities

Vulnerabilities are weaknesses in software that enable an attacker to compromise

the integrity, availability, or confidentiality of that software or the data it
processes. Some of the worst vulnerabilities allow attackers to run arbitrary code,
called exploits, on the compromised system. See Industry-Wide Vulnerability
Reports in the “Reference Guide” section of the Security Intelligence Report website
for more information about vulnerabilities.

Vulnerability Severity
The Common Vulnerability Scoring System (CVSS) is a standardized, platform-
independent scoring system for rating IT vulnerabilities. The CVSS assigns a
numeric value between 0 and 10 to vulnerabilities according to severity, with
higher scores representing greater severity. (See Vulnerability Severity in the
“Reference Guide” section of the Security Intelligence Report website for more
information.)

12
Figure 1. Industry-wide vulnerability disclosures by severity, 2006–2010

 Although the number of Medium and High severity vulnerabilities

disclosed is routinely much greater than the number of Low severity
vulnerability disclosures, the trend in 2010 is a positive one, with
Medium and High disclosures declining by 17.5 percent and 20.2
percent from 2009, respectively.

 Low severity vulnerability disclosures increased 45.8 percent, from 190

in 2009 to 277 in 2010.

 Mitigating the most severe vulnerabilities first is a security best practice.

High severity vulnerabilities that scored 9.9 or greater represent 5.5
percent of all vulnerabilities disclosed in 2010, as Figure 2 illustrates.
This percentage is down from 6.7 percent in 2009.

13
 Figure 2. Industry-wide vulnerability disclosures in 2010, by severity

Vulnerability Complexity
Some vulnerabilities are easier to exploit than others, and vulnerability
complexity is an important factor to consider in determining the magnitude of
the threat that a vulnerability poses. A High severity vulnerability that can only
be exploited under very specific and rare circumstances might require less
immediate attention than a lower severity vulnerability that can be exploited
more easily.

The CVSS gives each vulnerability a complexity ranking of Low, Medium, or

High. (See Vulnerability Complexity in the “Reference Guide” section of the
Security Intelligence Report website for more information about the CVSS
complexity ranking system.) Figure 3 shows the complexity mix for
vulnerabilities disclosed each year since 2006. Note that Low complexity
indicates greater danger, just as High severity indicates greater danger in Figure
2.

14
Figure 3. Industry-wide vulnerabilities by access complexity, 2006–2010

 As with vulnerability severity, the trend here is a positive one, with Low
and Medium complexity vulnerability disclosures declining 28.3 percent
and 5.0 percent from 2009, respectively.

 High complexity vulnerability disclosures increased 43.3 percent, from

120 in 2009 to 172 in 2010.

Operating System, Browser, and Application

Vulnerabilities
Figure 4 shows industry-wide vulnerabilities for operating systems, browsers,
and applications since 2006. (See Operating System, Browser, and Application
Vulnerabilities in the “Reference Guide” section of the Security Intelligence Report
website for an explanation of how operating system, browser, and application
vulnerabilities are distinguished.)

15
 Figure 4. Industry-wide operating system, browser, and application vulnerabilities, 2006–2010

 Application vulnerabilities continued to account for a large majority of all

vulnerabilities in 2010, although the total number of application
vulnerabilities declined 22.2 percent from 2009.

 Operating system and browser vulnerabilities remained relatively stable

by comparison, with each type accounting for a small fraction of the
total.

Vulnerability Disclosures
A disclosure, as the term is used in the SIR, is the revelation of a software
vulnerability to the public at large. It does not refer to any sort of private
disclosure or disclosure to a limited number of people. Disclosures can come
from a variety of sources, including the software vendor itself, security software
vendors, independent security researchers, and even malware creators.

The information in this section is compiled from vulnerability disclosure data

that is published in the National Vulnerability Database (http://nvd.nist.gov), the
U.S. government repository of standards-based vulnerability management.

16
Figure 5 charts vulnerability disclosures for Microsoft and non-Microsoft
products since 2006.

Figure 5. Vulnerability disclosures for Microsoft and non-Microsoft products, 2006–2010

 Vulnerability disclosures across the industry were down 16.5 percent in

2010 from 2009.

 This decline continues an overall trend of moderate declines since 2006.

This trend is likely because of better development practices and quality
control throughout the industry, which result in more secure software
and fewer vulnerabilities. (See Protecting Your Software in the “Managing
Risk” section of the Security Intelligence Report website for additional
details and guidance about secure development practices.)

 Vulnerability disclosures for Microsoft products increased slightly in

2010 but have generally remained stable over the past several periods.

 Vulnerabilities in Microsoft products accounted for 7.2 percent of all

vulnerabilities disclosed in 2010. This percentage is up from 4.5 percent
in 2009, primarily because of the overall decline in vulnerability
disclosures across the industry during that time.

17
 Exploits

An exploit is malicious code that takes advantage of software vulnerabilities to

infect a computer, without the user’s consent and usually without the user’s
knowledge. Exploits target vulnerabilities in the operating system, web browsers,
applications, or software components that are installed on the computer. In some
scenarios, targeted components are add-ons that are pre-installed by the
computer manufacturer before the computer is sold. A user may not even use the
vulnerable add-on or be aware that it is installed. Some software has no facility
for updating itself, so even if the software vendor publishes an update that fixes
the vulnerability, the user may not know that the update is available or how to
obtain it, and therefore remains vulnerable to attack.

Software vulnerabilities are enumerated and documented in the Common

Vulnerabilities and Exposures list (CVE) (http://cve.mitre.org), a standardized
repository of vulnerability information. Here and throughout this report, exploits
are labeled with the CVE identifier that pertains to the affected vulnerability, if
applicable. In addition, exploits that affect vulnerabilities in Microsoft software
are labeled with the Microsoft Security Bulletin number that pertains to the
vulnerability, if applicable.1

Figure 6 shows the prevalence of different types of exploits for each quarter in
2010.

1 See www.microsoft.com/technet/security/Current.aspx to search and read Microsoft Security Bulletins.

18
Figure 6. Exploits detected by Microsoft desktop antimalware products in 2010, by targeted platform or technology

 Malware written in Java has existed for many years, but attackers had not
focused significant attention on exploiting Java vulnerabilities until
somewhat recently. In 3Q10, the number of Java attacks increased to
fourteen times the number of attacks recorded in 2Q10, driven mostly
by the exploitation of a pair of vulnerabilities in versions of the Sun (now
Oracle) JVM, CVE-2008-5353 and CVE-2009-3867. Together, these two
vulnerabilities accounted for 85 percent of the Java exploits detected in
the second half of 2010.

 Exploits that target document editors and readers, such as Microsoft®

Word and Adobe Reader, declined in 2Q10 and remained at a lower
level thereafter.

 Operating system exploits, which have been less prevalent than other
types of exploits for several years, increased significantly in 3Q10,
primarily because of exploitation of two Windows® vulnerabilities.

19
 HTML and JScript/JavaScript Exploits
Figure 7 shows the prevalence of different types of HTML and
Jscript®/JavaScript exploits each quarter in 2010.

Figure 7. Types of HTML and JScript/JavaScript exploits detected by Microsoft desktop antimalware products in 2010

 Most of the exploits observed involved malicious HTML inline frames

(IFrames) that surreptitiously open pages hosting malicious code in
users’ web browsers.

 Exploits that target Windows Internet Explorer® vulnerabilities

accounted for between 19 and 36 percent of HTML-related exploits each
quarter. Most of these exploits targeted CVE-2010-0806, a vulnerability
that affects Internet Explorer versions 6 and 7 running on versions of
Windows earlier than Windows 7 and Windows Server 2008 R2.
Microsoft has issued Security Bulletin MS10-018 to address this
vulnerability. For more information, see the post “Active Exploitation of
CVE-2010-0806” (March 30, 2010) on the MMPC blog
(http://blogs.technet.com/mmpc).

20
Document Exploits
Figure 8 shows the prevalence of different types of document format exploits by
quarter in 2010.2

Figure 8. Types of document exploits detected by Microsoft desktop antimalware products in 2010

 Exploits that affected Adobe Acrobat and Adobe Reader accounted for
most document format exploits detected throughout 2010. Almost all of
these exploits involved the generic exploit family Win32/Pdfjsc.

 Adobe Acrobat and Adobe Reader exploits dropped by more than half
after the first quarter and remained near this reduced level throughout
the remainder of the year.

 Microsoft Office file format exploits accounted for between 0.5 and 2.8
percent of the document format exploits that were detected each quarter
in 2010.

2 Microsoft also detected a very small number of exploits that affect JustSystems Ichitaro, a Japanese-language
word processing program. These exploits affected fewer than 200 computers each quarter and are not shown in
the figure.

21
 Operating System Exploits
Figure 9 shows the prevalence of different operating system exploits by quarter in
2010.

Figure 9. Operating system exploits detected by Microsoft desktop antimalware products in 2010

 Several of the operating system exploits with the most detections in 2010
were caused by worms that spread in ways that result in large numbers
of detections on each computer they try to infect. Figure 9 provides
another perspective on these statistics, and shows the number of
individual computers that reported exploit attempts for several of these
exploits, in addition to the total number of detections.

 Operating system exploits had been declining for several years prior to
2010, and detections numbered less than 200,000 in each of the first two
quarters of the year. This decline changed in 3Q10 with the discovery
and publication of two zero-day exploits (exploits that take advantage of
undisclosed or newly disclosed vulnerabilities before the vendor releases
security updates for them) for two vulnerabilities that affect Windows,
CVE-2010-1885 and CVE-2010-2568.

 CVE-2010-1885 is a vulnerability that affects the Windows Help and

Support Center in Windows XP and Windows Server® 2003. Details of
the vulnerability were made public on June 10, 2010, about three weeks
before the end of the second quarter, and Microsoft issued an “out-of-
band” Security Bulletin, MS10-042, to address the vulnerability on July
13.

22
 Microsoft detected a relatively small number of exploits targeting CVE-
2010-1885 (fewer than 14,000 worldwide) in 2Q10, followed by a steep
rise to more than 250,000 detections in the third quarter. By the end of
the year, exploitation had declined significantly, with fewer than 65,000
detections in 4Q10.
For additional information, see the post Attacks on the Windows Help
and Support Center Vulnerability (CVE-2010-1885) (June 30, 2010) on
the MMPC blog, http://blogs.technet.com/mmpc.

 CVE-2010-2568 is a vulnerability that involves the way Windows Shell

handles shortcut files. This vulnerability was first discovered in mid-July
2010 following analysis of the Win32/Stuxnet worm, which uses the
vulnerability as a means of propagation. Microsoft issued an out-of-band
Security Bulletin, MS10-046, to address the vulnerability on August 2.
Initially, Stuxnet was the only family found to be making significant use
of CVE-2010-2568 exploits, but detections and removals rose as authors
of other malware families, including Win32/Vobfus and Win32/Sality,
began releasing new variants that exploited the vulnerability. For
additional information, see the post Stuxnet, malicious .LNKs, ...and
then there was Sality (July 30, 2010) on the MMPC blog,
http://blogs.technet.com/mmpc.

CVE-2010-2568 exploits affected about as many computers in 3Q10 as

CVE-2010-1885 exploits, but the number of detections per infected
computer was much higher (12.9 detections per infected computer,
compared to 1.5 for CVE-2010-1885). The Stuxnet worm uses USB
storage devices as its primary transmission vector, and the nature of the
shortcut vulnerability caused some computers to log large numbers of
detections as the Windows Shell repeatedly attempted to process the
same malicious shortcut file.

 CVE-2006-3439 is a vulnerability that affects the Server service in

Windows 2000, pre-Service Pack 3 versions of Windows XP, and pre-
Service Pack 2 versions of Windows Server 2003. Microsoft issued
Security Bulletin MS06-040 to address the vulnerability in August 2006.

In this case, although Microsoft detected significant numbers of infection

attempts targeting CVE-2006-3439, the actual number of computers
involved was quite small (fewer than 3,000 worldwide each quarter).
Exploits targeting network services, such as the Server service, can
generate large numbers of detections by real-time antimalware products:

23
 a worm traversing a network may make repeated attempts to infect an
individual computer using the exploit, with each unsuccessful attempt
logged as a separate detection.
In general, successful exploitation of operating system vulnerabilities as
old as CVE-2006-3439 should be rare, as most of the Windows
installations that were initially affected have since been updated with the
appropriate security updates or service packs or replaced by newer
versions of Windows that are not affected by the vulnerability. In 2010,
detections of CVE-2006-3439 exploits were strongly correlated with
detections of the uncommon Trojan family Win32/ServStart, suggesting a
possible connection between the two.

Security Breach Trends

In recent years, laws have been passed in a number of jurisdictions around the
world that require affected individuals to be notified when an organization loses
control of personally identifiable information (PII) with which it has been
entrusted. These mandatory notifications offer unique insights into how
information security efforts need to address issues of negligence as well as
technology.

The information in this section was generated from worldwide data security
breach reports from news media outlets and other information sources that
volunteers have recorded in the Data Loss Database (DataLossDB) at
http://datalossdb.org. (See Security Breach Trends in the “Reference Guide”
section of the Security Intelligence Report website for more information about the
DataLossDB and the breach types referenced here.)

24
Figure 10. Security breach incidents by incident type, 3Q09–4Q10

 The largest single category of incidents in each of the past six quarters
involved stolen equipment, ranging from a high of 34.5 percent of the
total in 3Q09 to a low of 18.6 percent of the total in 4Q10.

 Malicious incidents (those involving “hacking” incidents, malware, and

fraud) routinely account for less than half as many incidents as
negligence (involving lost, stolen, or missing equipment; accidental
disclosure; or improper disposal), as Figure 11 illustrates.

 Improper disposal of business records accounts for a significant portion

of incidents and is relatively easy for organizations to address by
developing and enforcing effective policies regarding the destruction of
paper and electronic records that contain sensitive information.

25
 Figure 11. Breach incidents resulting from attacks and negligence, 3Q09–4Q10

26
Malware and Potentially
Unwanted Software

Except where specified, the information in this section was compiled from
telemetry data that was generated from more than 600 million computers
worldwide and some of the busiest Internet online services. (See “Appendix B:
Data Sources” on page 71 for more information about the telemetry used in this
report.)

Global Infection Rates

The telemetry data generated by Microsoft® security products from users who
choose to opt in to data collection includes information about the location of the
computer, as determined by the setting of the Location tab or menu in Regional
and Language Options in Control Panel. This data makes it possible to compare
infection rates, patterns, and trends in different locations around the world.
Figure 12. The locations with the most computers reporting detections and removals by Microsoft desktop antimalware
products in 2010

Chg. 3Q
Country/Region 1Q10 2Q10 3Q10 4Q10
to 4Q
1 United States 11,025,811 9,609,215 11,340,751 11,817,437 4.2% ▲
-2.1%
2 Brazil 2,026,578 2,354,709 2,985,999 2,922,695
▼
-8.6%
3 China 2,168,810 1,943,154 2,059,052 1,882,460
▼
18.9%
5 United Kingdom 1,490,594 1,285,570 1,563,102 1,857,905
▲
12.1%
4 France 1,943,841 1,510,857 1,601,786 1,794,953
▲
56.8%
7 Korea 962,624 1,015,173 1,070,163 1,678,368
▲
-3.9%
6 Spain 1,358,584 1,348,683 1,588,712 1,526,491
▼

27
 Chg. 3Q
Country/Region 1Q10 2Q10 3Q10 4Q10
to 4Q
41.3%
9 Russia 700,685 783,210 928,066 1,311,665
▲
10.6%
8 Germany 949,625 925,332 1,177,414 1,302,406
▲
10.8%
10 Italy 836,593 794,099 900,964 998,458
▲

 Detections in Korea rose 56.8 percent from 3Q10 to 4Q10, with three
families—Win32/Onescan, Win32/Parite, and Win32/Nbar—
representing 77 percent of the 3Q–4Q increase. Onescan, a Korean-
language rogue security software family first detected in 4Q10, was itself
responsible for about 32 percent of all detections in Korea. (For more
information, see “Rogue Security Software” on page 41.)
Figure 13. False malware detections by Win32/Onescan, a Korean-language rogue security software family

28
  Detections in Russia rose 41.3 percent from 3Q to 4Q, primarily because
of a significant increase in the number of computers running Microsoft
Security Essentials there.

In absolute terms, the locations with the most computers reporting detections
tend to be ones with large populations and large numbers of computers. To
control for this effect, Figure 14 shows the infection rates in locations around the
world using a metric called computers cleaned per mille (thousand), or CCM,
which represents the number of reported computers cleaned in a quarter for
every 1,000 executions of the Microsoft Windows® Malicious Software Removal
Tool (MSRT).3 (See the Security Intelligence Report website for more information
about the CCM metric.)
Figure 14. Infection rates by country/region in 1H10 (top) and 2H10 (bottom), by CCM

3 For the maps in Figure 14, the CCM totals are averaged for the first two and last two quarters of 2010,
respectively, to produce CCM totals for 1H10 and 2H10.

29
  Among locations with at least 100,000 executions of MSRT in 4Q10,
Korea had the highest infection rate, with 40.3 computers cleaned for
every 1,000 MSRT executions (CCM 40.3). Following Korea were Spain
(33.2), Turkey (32.8), Taiwan (24.3), and Brazil (20.8).

 For the entire year, Turkey had the highest average quarterly CCM at
36.8, followed by Spain (36.1), Korea (34.8), Taiwan (29.7), and Brazil
(24.7). These five locations have consistently had the highest infection
rates among large countries and regions for most of the past six quarters,
as shown in Figure 15 on page 31.

 Locations with low infection rates include Mongolia (1.3 average CCM
for 2010), Bangladesh (1.4), and Belarus (1.6). Large countries and
regions with low infection rates include the Philippines (3.1), Austria
(3.4), India (3.8), and Japan (4.4).

Detections and removals in individual countries/regions can vary significantly

from period to period. Increases in the number of computers with detections can
be caused not only by increased prevalence of malware in that country but also
by improvements in the ability of Microsoft antimalware solutions to detect
malware. Large numbers of new antimalware installations in a location also
typically increase the number of computers cleaned in that location.

The next two figures illustrate infection rate trends for specific locations around
the world, relative to the trends for all locations with at least 100,000 MSRT
executions each quarter in 2010. (See Infection Trends Worldwide in the “Key

30
Findings” section of the Security Intelligence Report website for additional details
about this information.)

Figure 15. Trends for the five locations with the highest infection rates in 4Q10, by CCM (100,000 MSRT executions
minimum per quarter in 2010)

 Korea has come under sustained attack in recent quarters, resulting in a

dramatic rise from 4th place in 3Q10 to 1st place in 4Q10. The CCM in
Korea rose from 23.6 in 4Q09 to 40.3 a year later, an increase of 16.7
points, or 71.1 percent—the largest such increase over the past year. (See
the “Global Threat Assessment” section of the Security Intelligence Report
website for more information about threats in Korea.)

 Korea, Spain, Turkey, Taiwan, and Brazil have occupied the top five
spots among large countries and regions with the highest infection rates
in all but one of the last six quarters (the sole exception being 4Q09,
when Portugal edged Korea for 5th place).

31
 Figure 16. Infection rate trends for the five most improved locations between 4Q09 and 4Q10, by CCM (100,000 MSRT
executions minimum in 4Q10)

 The most improved locations are those that showed the greatest decline
in CCM between 4Q09 and 4Q10.

 Brazil, though still one of the locations with the highest infection rates,
has improved significantly over the past six quarters, dropping from 30.1
CCM in 3Q09 to 20.8 in 4Q10. Declines in Win32/Frethog and
Win32/Hamweq were chiefly responsible for this improvement, followed
by declines in Win32/Conficker and Win32/Rimecud. (See “Threat
Families” on page 39 for more information about these and other
malware families.)

 Although the total number of detections and removals in Russia

increased through 2010, as explained on page 28, the actual infection
rate declined significantly, from 17.3 CCM in 3Q09 to 10.1 in 4Q10.
This decrease was primarily because of decreases in Conficker, Hamweq,
and Win32/Taterf.

 Infection rates in Portugal and Bahrain fluctuated over the past six
quarters, but both locations ended 4Q10 showing significant
improvements over 3Q09. Portugal went from 25.0 CCM to 15.6, a 37.6

32
 percent decrease. Bahrain dropped from 13.6 to 9.0, a decline of 33.8
percent.

 The CCM in China decreased from 9.5 in 3Q09 to 2.9 in 4Q10.

Although this makes China one of the locations with the lowest infection
rates worldwide as measured by CCM, a number of factors that are
unique to China are important to consider when assessing the state of
computer security there. The malware ecosystem in China is dominated
by a number of Chinese- language threats that are not prevalent
anywhere else. The CCM figures are calculated based on telemetry from
MSRT, which targets global malware families. To date, we have not
targeted families specific to China with. In 2010, for example, 92 to 94
percent of the threats reported by computers running Microsoft Security
Essentials in China would not have been detected by MSRT. For a more
in-depth perspective on the threat landscape in China, see the “Global
Threat Assessment” section of the Security Intelligence Report website.

Operating System Infection Rates

The features and updates that are available with different versions of the
Windows operating system, along with the differences in the way people and
organizations use each version, affect the infection rates for the different versions
and service packs. Figure 17 shows the infection rate for each Windows
operating system/service pack combination that accounted for at least 0.1 percent
of total MSRT executions in 2010.

33
 Figure 17. Average quarterly infection rate (CCM) by operating system and service pack in 2010

“32” = 32-bit; “64” = 64-bit. Supported systems with at least 0.1 percent of total executions shown.

 This data is normalized: the infection rate for each version of Windows is
calculated by comparing an equal number of computers per version (for
example, 1,000 Windows XP SP2 computers to 1,000 Windows 7 RTM
computers).

 As in previous periods, infection rates for more recently released

operating systems and service packs are consistently lower than earlier
ones, for both client and server platforms. Windows 7 and Windows
Server® 2008 R2, the most recently released Windows client and server
versions, respectively, have the lowest infection rates on the chart.

 Infection rates for the 64-bit versions of Windows Vista® and Windows
7 are lower than for the corresponding 32-bit versions of those operating
systems. One reason may be that 64-bit versions of Windows still appeal
to a more technically savvy audience than their 32-bit counterparts,
despite increasing sales of 64-bit Windows versions among the general
computing population. Kernel Patch Protection (KPP), a feature of 64-bit
versions of Windows that protects the kernel from unauthorized
modification, may also contribute to the discrepancy by preventing
certain types of malware from operating.

34
Figure 18. CCM trends for supported 32-bit versions of Windows XP, Windows Vista, and Windows 7, 3Q09-4Q10

 As Figure 18 shows, Windows 7 has consistently had the lowest infection

rate of any 32-bit client operating system/service pack combination over
the past six quarters.

35
 Threat Categories
The Microsoft Malware Protection Center (MMPC) classifies individual threats
into types based on a number of factors, including how the threat spreads and
what it is designed to do. To simplify the presentation of this information and
make it easier to understand, the SIR groups these types into 10 categories based
on similarities in function and purpose.

Figure 19. Detections by threat category each quarter in 2010, by percentage of all computers reporting detections

Round markers indicate malware categories; square markers indicate potentially unwanted software categories.

 Totals for each time period may exceed 100 percent because some
computers have more than one category of threat detected and removed
from them in each time period.

 The miscellaneous trojans category, which consists of all trojans that are
not categorized as trojan downloaders & droppers, was the most
prevalent category each quarter in 2010, with detections on 20.0 percent
of all infected computers in 4Q10, down from 22.7 percent in 1Q10.

 Detections of adware increased significantly during the second half of the

year, rising from 8.9 percent of infected computers in 2Q10 to 15.1
percent in 4Q10. This increase was almost completely caused by the
appearance of a pair of new adware families, JS/Pornpop and

36
 Win32/ClickPotato, in the third quarter. (See “Threat Families” on page
39 for more information about these and other families.)

 After increasing from 1Q10 to 2Q10, worms declined significantly

through the end of the year, from a second-quarter high of 19.2 percent
of infected computers to 13.5 percent in 4Q10. A 61.3 percent decrease
in detections and removals of the worm family Win32/Hamweq between
1Q10 and 4Q10 is partially responsible for this relative decline,
combined with increases in other categories. (Hamweq was added to
MSRT in December 2009, and was detected by the tool on more than 1
million computers by the end of 1Q10. By the end of the year, Hamweq
detections had declined significantly, with MSRT removing it from fewer
than 300,000 computers in 4Q10.)

 The miscellaneous potentially unwanted software and trojan

downloaders & droppers categories began the year at similar levels of
prevalence, and then diverged. miscellaneous potentially unwanted
software rose from 16.1 percent of infected computers to 18.1 percent,
with increased detections of the potentially unwanted software families
Win32/Zwangi and Win32/Keygen accounting for much of the increase
(the increase in detections of the latter family was caused more by
improved detection than by increased prevalence). trojan downloaders &
droppers declined from 14.7 percent to 11.6 percent, in part because of
a decline in detections of Win32/Renos, a perennially common family.

 Each of the other categories was detected on fewer than 10 percent of

infected computers. Password Stealers & Monitoring Tools declined to
6.6 percent of infected computers in 4Q10 following a decrease in
detections of Win32/Frethog, which targets passwords for online games.
Spyware, which has never been very common, declined even more in
2010 to just 0.2 percent of infected computers in the fourth quarter.

Threat Categories by Location

There are significant differences in the types of threats that affect users in
different parts of the world. The spread and effectiveness of malware are highly
dependent on language and cultural factors, in addition to the methods used for
distribution. Some threats are spread using techniques that target people who
speak a particular language or who use services that are local to a particular
geographic region. Other threats target vulnerabilities or operating system
configurations and applications that are unequally distributed around the globe.

37
 Figure 20 shows the relative prevalence of different categories of malware and
potentially unwanted software in several locations around the world in 2010.

Figure 20. Threat category prevalence worldwide and in nine individual locations in 2010

Russi
Category World US Brazil China UK Fr. Spain Ger. Korea
a
43.4 23.2 28.0 36.5 21.6 20.1 40.3 28.4 17.3
Misc. Trojans 31.6%
% % % % % % % % %
Misc. Potentially 22.6 31.2 52.1 23.6 24.3 22.6 33.8 24.5 10.3
25.5%
Unwanted Software % % % % % % % % %
16.6 35.6 13.5 11.8 21.0 40.2 32.8 14.4 40.1
Worms 24.4%
% % % % % % % % %
Trojan Downloaders 20.2 26.2 18.8 20.3 19.7 16.9 17.0 28.9
20.1% 8.0%
& Droppers % % % % % % % %
21.4 29.3 33.0 10.7 16.3 12.1
Adware 17.4% 9.4% 3.4% 8.2%
% % % % % %
Password Stealers 27.9 10.7 20.5 10.3 14.7
11.7% 6.1% 7.5% 9.2% 9.3%
& Monitoring Tools % % % % %
10.5 13.5
Exploits 7.1% 9.6% 7.3% 2.7% 3.0% 8.0% 5.7% 3.3%
% %
10.3
Backdoors 6.6% 5.3% 5.7% 4.2% 4.4% 8.4% 8.2% 5.1% 7.1%
%
10.3 12.1 13.8
Viruses 5.9% 5.1% 6.1% 3.4% 3.3% 3.7% 3.2%
% % %

Spyware 0.6% 0.7% 0.2% 2.3% 0.4% 0.3% 0.2% 0.5% 0.7% 0.5%
Totals for each location exceed 100 percent because some computers reported threats from more than one category.

 Within each row of Figure 20, a darker color indicates that the category
is more prevalent in the specified location than in the others, and a
lighter color indicates that the category is less prevalent.

 The United States and the United Kingdom, two predominantly English-
speaking locations that also share a number of other cultural similarities,
have similar threat mixes in most categories. Exceptions include Adware,
which is more common in the UK, and Worms, which are more
common in the US.

 Brazil has an unusually high concentration of Password Stealers &

Monitoring Tools, primarily because of the prevalence of Win32/Bancos,
which targets customers of Brazilian banks.

38
  China has a relatively high concentration of Miscellaneous Potentially
Unwanted Software, Exploits, Backdoors, and Spyware, and a relatively
low concentration of Worms and Adware. China routinely exhibits a
threat mix that is much different than those of other large countries and
regions. Two of the most common threats in China, Win32/BaiduSobar
and Win32/Sogou, are Chinese-language potentially unwanted software
families that are uncommon elsewhere. The most common families in
China also include a pair of exploits, JS/CVE-2010-0806 and
JS/ShellCode, that were less prevalent elsewhere.

 Adware dominates in France, led by Win32/ClickPotato.

 Worms and Backdoors are unusually common in Spain. The top six
families detected in Spain in 2010 were worms.

 The threat mix in Russia resembles that of the world as a whole, with the
exception of an unusually low concentration of Adware, perhaps because
of the highly language-dependent nature of online advertising.

 In Germany, Trojan Downloaders & Droppers are nearly twice as

common as in the rest of the world, led by Win32/Renos.

 Korea has a large concentration of viruses, led by Win32/Parite, and

worms. Viruses and worms have long been unusually common in Korea,
perhaps because of the popularity of public Internet gaming centers there
where viruses are easily transmitted between computers and removable
volumes.

See “Appendix C: Worldwide Infection Rates” on page 73 for more information

about malware around the world.

Threat Families
Figure 21 lists the top 10 malware and potentially unwanted software families
that were detected on computers by Microsoft desktop security products in the
second half of 2010.

39
 Figure 21. Quarterly trends for the top 10 malware and potentially unwanted software families detected by Microsoft
desktop security products in 2H10

Family Most Significant Category 1Q10 2Q10 3Q10 4Q10

1 JS/Pornpop Adware — — 2,660,061 3,860,365
2 Win32/Autorun Worms 1,256,649 1,646,532 2,805,585 3,314,092
3 Win32/Taterf Worms 1,496,780 2,323,750 2,338,517 1,615,649
Misc. Potentially Unwanted
4 Win32/Zwangi 542,534 860,747 1,638,398 2,299,210
Software
Trojan Downloaders &
5 Win32/Renos 2,693,093 1,889,680 2,109,631 1,655,865
Droppers
6 Win32/Rimecud Worms 1,809,231 1,749,708 1,674,975 1,892,919
7 Win32/Conficker Worms 1,498,256 1,664,941 1,649,934 1,744,986
8 Win32/FakeSpypro Miscellaneous Trojans 1,244,903 1,424,152 1,897,420 889,277
9 Win32/Hotbar Adware 1,015,659 1,483,289 942,281 1,640,238
10 Win32/ClickPotato Adware — — 451,660 2,110,117

Figure 22. The families that increased the most in prevalence in 2010

 JS/Pornpop, the most commonly detected family in 4Q10, is a detection

for specially crafted JavaScript-enabled objects that attempt to display
pop-under advertisements in users’ web browsers, usually with adult
content.

40
 Pornpop is one of the fastest spreading malware families seen in several
years. First detected in August 2010, it quickly grew to become the
second most prevalent family in 3Q10, and the most prevalent family in
4Q10 and in the second half of the year as a whole.

 Detections and removals of Win32/Autorun, a generic detection for

worms that spread between mounted volumes using the Autorun feature
of Windows, increased significantly in 4Q10, although Autorun dropped
to second place because of the spread of Pornpop.

 Win32/Taterf, the most prevalent threat in 2Q10, dropped to third by

4Q10. Taterf belongs to a category of threats that are designed to steal
passwords for popular online computer games and transmit them to the
attackers. See “Online Gaming-Related Families” on page 62 of Microsoft
Security Intelligence Report, Volume 5 (January through June 2008) for more
information about these threats.

 Win32/Renos, the most prevalent threat in 1Q10, dropped to fifth by

4Q10. Renos is a family of Trojan downloaders that is often used to
install rogue security software. Since 2006, it has consistently been one
of the threats most commonly detected and removed by Microsoft
antimalware desktop products and services.

 The potentially unwanted software family Win32/Zwangi rose from tenth

in 2Q10 to fourth in 4Q10. Zwangi is a program that runs as a service in
the background and modifies web browser settings to visit a particular
website.

 The adware family Win32/ClickPotato, first detected in August 2010,

rose quickly to become the tenth most prevalent family in 4Q10.
ClickPotato is a program that displays pop-up and notification-style
advertisements based on the user’s browsing habits.

Rogue Security Software

Rogue security software has become one of the most common methods that
attackers use to swindle money from victims. Rogue security software, also
known as scareware, is software that appears to be beneficial from a security
perspective but provides limited or no security, generates erroneous or
misleading alerts, or attempts to lure users into participating in fraudulent
transactions. These programs typically mimic the general look and feel of

41
 legitimate security software programs and claim to detect a large number of
nonexistent threats while urging users to pay for the “full version” of the software
to remove the threats. Attackers typically install rogue security software programs
through exploits or other malware or use social engineering to trick users into
believing the programs are legitimate and useful. Some versions emulate the
appearance of the Windows Security Center or unlawfully use trademarks and
icons to misrepresent themselves. (See Rogue Security Software in the “Reference
Guide” section of the Security Intelligence Report website for more information
about this kind of threat. Also see
www.microsoft.com/security/antivirus/rogue.aspx for an informative series of
videos about rogue security software aimed at a general audience.)
Figure 23. Some of the “brands” used by different versions of the rogue security software family
Win32/FakeXPA

Figure 24 shows detection trends for the most common rogue security software
families detected in 2010.

42
Figure 24. Trends for the most commonly detected rogue security software families in 2010, by quarter

 Win32/FakeSpypro was the most commonly detected rogue security

software family in each quarter of 2010, with more than twice as many
detections and removals overall as the next most prevalent family. Names
under which FakeSpypro is distributed include AntispywareSoft,
Spyware Protect 2009, and Antivirus System PRO. Detections for
FakeSpypro were added to MSRT in July 2009.

 Win32/FakeXPA, the second most commonly detected rogue security

software family overall in 2010, fell from a near tie with FakeSpypro in
1Q10 to sixth place in 4Q10. FakeXPA is a persistent, frequently
updated threat that uses a variety of techniques to evade detection and
removal by legitimate security products. It is distributed under a large
number of names, some of which are shown in Figure 23. Detections for
FakeXPA were added to MSRT in December 2008.

 Win32/FakePAV was first detected in 3Q10 and rose quickly to become

the second most commonly detected rogue security software family in
the fourth quarter. FakePAV is one of several rogue security software
families that masquerade as Microsoft Security Essentials. It presents a
dialog box that is similar in appearance to a Security Essentials alert,
listing one or more nonexistent infections that it claims it cannot remove.
It then offers to “install” a trial version of a different security program

43
 (actually another part of FakePAV itself), after which it proceeds in a
manner similar to other rogue security software programs.

Figure 25. A genuine Microsoft Security Essentials alert (top) and a fake alert generated by
Win32/FakePAV (bottom)

Names under which FakePAV is distributed include Red Cross Antivirus,

Peak Protection 2010, AntiSpy Safeguard, Major Defense Kit, Pest
Detector, ThinkPoint, Privacy Guard 2010, Palladium Pro, and others.
Detections for FakePAV were added to MSRT in November 2010. For
additional information, see the post MSRT Tackles Fake Microsoft
Security Essentials (November 9, 2010) on the MMPC blog,
http://blogs.technet.com/mmpc.

44
Home and Enterprise Threats
The usage patterns of home users and enterprise users tend to be very different.
Enterprise users typically use computers to perform business functions while
connected to a network, and may have limitations placed on their Internet and
email usage. Home users are more likely to connect to the Internet directly or
through a home router and to use their computers for entertainment purposes,
such as playing games, watching videos, and communicating with friends. These
different usage patterns mean that home users tend to be exposed to a different
mix of computer threats than enterprise users.

The infection telemetry produced by Microsoft desktop antimalware products

and tools includes information about whether the infected computer belongs to
an Active Directory® Domain Services domain. Domains are used almost
exclusively in enterprise environments, and computers that do not belong to a
domain are more likely to be used at home or in other non-enterprise contexts.
Comparing the threats encountered by domain computers and non-domain
computers can provide insights into the different ways attackers target enterprise
and home users and which threats are more likely to succeed in each
environment.

Figure 26 and Figure 27 list the top 10 families detected on domain-joined and
non-domain computers in 4Q10.

45
 Figure 26. Top 10 families detected on domain-joined computers in 2010, by percentage of domain-joined computers
reporting detections

Family Most Significant Category 1Q10 2Q10 3Q10 4Q10

1 Win32/Conficker Worms 21.3% 22.0% 19.6% 18.9%
2 Win32/Autorun Worms 7.3% 8.3% 10.0% 10.0%
3 Win32/Rimecud Worms 9.0% 9.8% 8.0% 8.3%
4 Win32/Taterf Worms 4.1% 6.9% 5.9% 4.1%
Miscellaneous Potentially Unwanted
5 Win32/RealVNC 5.6% 5.4% 4.9% 4.3%
Software
6 Win32/Hamweq Worms 7.0% 5.3% 3.2% 2.4%
7 Win32/Frethog Password Stealers & Monitoring Tools 6.5% 6.0% 2.8% 2.4%
8 Win32/Renos Trojan Downloaders & Droppers 5.2% 3.4% 4.0% 2.8%
9 Win32/Alureon Miscellaneous Trojans 2.7% 2.4% 2.8% 1.8%
10 Win32/FakeSpypro Miscellaneous Trojans 2.3% 3.0% 2.8% 0.9%

46
Figure 27. Top 10 families detected on non-domain computers in 2010, by percentage of all infected non-domain
computers reporting detections

Family Category Q1 Q2 Q3 Q4
1 Win32/Renos Trojan Downloaders & Droppers 8.8% 6.6% 6.1% 4.6%
2 Win32/Autorun Worms 3.8% 5.4% 7.8% 8.7%
3 Win32/Taterf Worms 4.8% 8.0% 6.7% 4.4%
4 Win32/Rimecud Worms 5.6% 5.7% 4.6% 5.0%
5 Win32/Frethog Password Stealers & Monitoring Tools 6.4% 6.9% 3.6% 3.4%
6 JS/Pornpop Adware — — 7.8% 10.4%
7 Win32/FakeSpypro Miscellaneous Trojans 4.1% 4.9% 5.6% 2.5%
8 Win32/Conficker Worms 3.8% 4.7% 3.9% 3.8%
9 Win32/Zwangi Miscellaneous Potentially Unwanted Software 1.8% 3.1% 4.9% 6.4%
1
Win32/Hotbar Adware 3.4% 5.3% 2.8% 4.6%
0

 Seven families are common to both lists, although they are ordered
differently and in different proportions. The worm family
Win32/Conficker, which uses several methods of propagation that work
more effectively within a typical enterprise network environment than
they do over the public Internet, leads the domain-joined list by a
significant margin, but ranks ninth on the non-domain list.

47
  Worms accounted for five of the top 10 families detected on domain-
joined computers. Several of these worms, including Conficker,
Win32/Autorun, and Win32/Taterf, are designed to propagate via
network shares, which are common in domain environments.

 On non-domain computers, JS/Pornpop was the most commonly

detected family in 4Q10 and the fourth most commonly detected family
in 2010 overall. By contrast, this family was detected much less often on
domain-joined computers. Pornpop is an adware family that attempts to
display pop-under advertisements that usually contain adult content in
users’ web browsers. It is often found on websites that host illegal or
illicit content, which users in domain environments are often restricted
from accessing by organizational policy or blocking software.

 Taterf and Win32/Frethog are two related families that are designed to
steal the passwords of users who play massively multiplayer online role-
playing games (MMORPGs). Such games are not common in the
workplace, yet both families were detected with similar frequency on
both domain-joined and non-domain computers. Taterf and Frethog
both rely heavily on removable drives to propagate—a technique that
was probably developed to help spread them in Internet cafés and public
gaming centers, but one that has had the effect of spreading them
efficiently in enterprise environments as well, which was perhaps
unexpected.

48
 Email Threats

Most of the email messages sent over the Internet are unwanted. Not only does
all this unwanted email tax recipients’ inboxes and the resources of email
providers, but it also creates an environment in which emailed malware attacks
and phishing attempts can proliferate. Email providers, social networks, and
other online communities have made blocking spam, phishing, and other email
threats a top priority.

Spam Messages Blocked

The information in this section is compiled from telemetry data provided by
Microsoft Forefront® Online Protection for Exchange (FOPE), which provides
spam, phishing, and malware filtering services for thousands of enterprise
customers and tens of billions of messages per month. (See Spam Trends in the
“Reference Guide” section of the Security Intelligence Report website for more
information.)
Figure 28. Messages blocked by FOPE each month in 2010

49
  After increasing gradually and then reaching a plateau through the first
eight months of 2010, the number of spam messages received and
blocked by FOPE dropped abruptly in September, and again in
December. These drops can be correlated with events involving two of
the world’s most significant spam-sending botnets:

o During the last week of August, researchers affiliated with the

security firm LastLine spearheaded a coordinated takedown of
command-and-control (C&C) servers associated with the
Win32/Cutwail spambot. In the days following the takedown, FOPE
recorded a significant drop in the average daily volume of messages
blocked.

o Around December 25, spam researchers around the world recorded

an almost complete cessation of spam originating from the large
Rustock botnet, with some spam trackers reporting a drop in the
global spam rate as high as 50 percent or more. During the final
week of December, the number of messages blocked by FOPE was
almost 30 percent less than in the prior week, compared to a drop of
less than two percent between the final two weeks of 2009. The
Rustock botnet subsequently began sending spam again in mid-
January, and the number of messages blocked by FOPE has risen
accordingly. The reasons for this hiatus are still being investigated.

FOPE performs spam filtering in two stages. Most spam is blocked by servers at
the network edge, which use reputation filtering and other non-content-based
rules to block spam or other unwanted messages. Messages that are not blocked
at the first stage are scanned using content-based rules, which detect and filter
many additional email threats, including attachments that contain malware.

50
Figure 29. Percentage of incoming messages blocked by FOPE using edge-blocking and content filtering in 2010

 In 2010 overall, only about one out of every 38.5 incoming messages
made it to recipients’ inboxes. The rest were blocked at the network edge
or through content filtering.

 Approximately 95.3 percent of all incoming messages were blocked at

the network edge, which means that only 4.7 percent of incoming
messages had to be subjected to the more resource-intensive content
filtering process.

 The effectiveness of edge-filtering techniques such as IP address

reputation checking, SMTP connection analysis, and recipient validation
have increased dramatically over the past several years, which enables
mail-filtering services to provide better protection to users even as the
total amount of unwanted message traffic on the Internet remains very
high.

Spam Types
The FOPE content filters recognize several different common types of spam
messages. Figure 30 shows the relative prevalence of these spam types in 2010.

51
 Figure 30. Inbound messages blocked by FOPE filters in 2010, by category

 Advertisements for nonsexual pharmaceutical products accounted for

32.4 percent of the spam messages blocked by FOPE content filters in
2010.

 Together with nonpharmaceutical product ads (18.3 percent of the total)

and advertisements for sexual performance products (3.3 percent),
product advertisements accounted for 54.0 percent of spam in 2010,
which is down from 69.2 percent a year ago.

 In an effort to evade content filters, spammers often send messages that

consist only of one or more images, with no text in the body of the
message. Image-only spam messages accounted for 8.7 percent of the
total in 2010, up from 6.3 percent in 2009.

52
Figure 31. Inbound messages blocked by FOPE content filters each month in 2010, by category

53
  Nonsexual pharmaceutical ads and nonpharmaceutical product ads were
the most highly ranked categories by a significant margin throughout
most of 2010.

 As Figure 31 illustrates, spam categories can vary considerably from

month to month as spammers conduct time-based campaigns, much like
legitimate advertisers do. Spam that advertises fraudulent university
diplomas, typically a low-volume category, increased nearly six fold
between February and March and was actually the third most prevalent
category in March and April before declining to last place in June.
Similarly, image-only ads, which accounted for a small and declining
percentage of spam through May, suddenly began rising in prominence
in June, briefly eclipsed nonpharmaceutical product ads in August, and
then returned to more typical levels through the end of the year.

54
Malicious Websites

Attackers often use websites to conduct phishing attacks or distribute malware.

Malicious websites typically appear completely legitimate and often provide no
outward indicators of their malicious nature, even to experienced computer
users. To help protect users from malicious webpages, Microsoft and other
browser vendors have developed filters that keep track of sites that host malware
and phishing attacks and display prominent warnings when users try to navigate
to them.

The information in this section is compiled from a variety of internal and external
sources, including telemetry data produced by the SmartScreen® filter (in
Windows® Internet Explorer® 8 and 9), the Phishing Filter (in Internet Explorer
7), from a database of known active phishing and malware hosting sites reported
by users of Internet Explorer and other Microsoft® products and services, and
from malware data provided by Microsoft antimalware technologies. (See
Phishing and Malware Hosts in the “Reference Guide” section of the Security
Intelligence Report website for more information.)

55
 Figure 32. The SmartScreen filter in Internet Explorer 8 and 9 blocks reported phishing and malware
distribution sites

Phishing Sites
Figure 33 compares the volume of active phishing sites in the SmartScreen
database each month with the volume of phishing impressions tracked by Internet
Explorer. A phishing impression is a single instance of a user attempting to visit a
known phishing site with Internet Explorer and being blocked.

56
Figure 33. Phishing sites and impressions tracked each month in 2010, relative to the monthly average for each

 Sudden sharp spikes in impressions like the one shown in June are not
unusual. Phishers often engage in discrete campaigns that are intended to
drive more traffic to each phishing page, without necessarily increasing
the total number of active phishing pages they are maintaining at the
same time. In this case, the June increase is not strongly correlated with
increases in any particular type of target institution.

 Phishing impressions and active phishing pages rarely correlate strongly

with each other. The total number of active phishing pages tracked by
Microsoft remained very stable from month to month, with no month
deviating by more than about 15 percent from the six-month average.

Target Institutions
Figure 34 and Figure 35 show the percentage of phishing impressions and active
phishing sites, respectively, recorded by Microsoft during each month in 2010
for the most frequently targeted types of institutions.

57
 Figure 34. Impressions for each type of phishing site each month in 2010

Figure 35. Active phishing sites tracked each month in 2010, by type of target

 Phishers have traditionally targeted financial sites more than other types
of sites, but 2010 showed evidence of a shift to social networks. Phishing
impressions that targeted social networks increased from a low of 8.3
percent of all impressions in January to a high of 84.5 percent of
impressions in December. In particular, the final four months of the year

58
 show signs of a strong and sustained phishing campaign or campaigns
against social networks.

 Early in 2010, phishers showed signs of targeting online gaming sites

with increased frequency, although this push appears to have dwindled
as social networks came under increased attack. Impressions that
targeted gaming sites reached a high of 16.7 percent of all impressions in
June before dropping to a more typical 2.1 percent in December.

 Phishing sites that target social networks routinely receive the highest
number of impressions per active phishing site. The percentage of active
phishing sites that targeted social networks increased during the final
months of the year, but still only accounted for 4.2 percent of active sites
in December, despite receiving 84.5 percent of impressions that month.
Nevertheless, the number of active sites targeting gaming sites remained
relatively high during the second half of the year, which suggests that
more campaigns may be coming.

 As in previous periods, phishing sites that targeted financial institutions

accounted for the majority of active phishing sites, ranging from 78 to 91
percent of sites each month. Financial institutions targeted by phishers
can number in the hundreds, and customized phishing approaches are
required for each one. By contrast, just a handful of popular sites account
for the bulk of the social network and online service usage on the
Internet, so phishers can effectively target many more people per site.
Still, the potential for direct illicit access to victims’ bank accounts means
that financial institutions remain perennially popular phishing targets,
and they continue to receive the largest or second-largest number of
impressions each month.

Global Distribution of Phishing Sites

Phishing sites are hosted all over the world on free hosting sites, on compromised
web servers, and in numerous other contexts. Performing geographic lookups of
IP addresses in the database of reported phishing sites makes it possible to create
maps that show the geographic distribution of sites and to analyze patterns.

59
 Figure 36. Phishing sites per 1,000 Internet hosts for locations around the world in 1H10 (top) and 2H10 (bottom)

 The worldwide distribution of phishing sites remained largely consistent

between the first and second halves of the year.

 Phishing sites are concentrated in a few locations but have been detected
on every inhabited continent.

 Locations with smaller populations and fewer Internet hosts tend to have
higher concentrations of phishing pages, although in absolute terms most

60
 phishing pages are located in large, industrialized countries/regions with
large numbers of Internet hosts.

Malware Hosting Sites

The SmartScreen filter in Internet Explorer 8 and 9 helps provide protection
against sites that are known to host malware, in addition to phishing sites. The
SmartScreen antimalware feature uses URL reputation data and Microsoft
antimalware technologies to determine whether those servers distribute unsafe
content. As with phishing sites, Microsoft keeps track of how many people visit
each malware hosting site and uses the information to improve the SmartScreen
filter and to better combat malware distribution. (See Malware Hosts in the
“Reference Guide” section of the Security Intelligence Report website for more
information.)
Figure 37. The SmartScreen filter in Internet Explorer 8 (top) and Internet Explorer 9 (bottom)
displays a warning when a user attempts to download an unsafe file

Figure 38 compares the volume of active malware hosting sites in the

SmartScreen database each month with the volume of malware impressions
tracked by Internet Explorer.

61
 Figure 38. Malware hosting sites and impressions tracked each month in 2010, relative to the monthly average for each

 The number of active malware hosting sites tracked each month

increased gradually through the year, mostly because of improved
detection.

 After a rising trend during the first five months, the number of malware
hosting impressions decreased each month for the rest of the year.
Malware host protection in browsers is a relatively new development
compared to phishing protection, and it is possible that attackers are
reacting by moving away from this method of distribution to other
techniques.

Malware Categories
Figure 39 and Figure 40 show the types of threats hosted at URLs that were
blocked by the SmartScreen filter in 2H10.

62
Figure 39. Threats hosted at URLs blocked by the SmartScreen filter in 2010, by category

63
 Figure 40. The top 10 malware families hosted on sites blocked by the SmartScreen filter in 1H10 and 2H10, by percent of
all such sites

1H10 Most Significant 2H10 Most Significant

Threat Name Percent Threat Name Percent
Rank Category Rank Category

Misc. Potentially Misc. Potentially

1 Win32/MoneyTree 61.1 1 Win32/MoneyTree 47.3
Unwanted Software Unwanted Software
Miscellaneous Trojan Downloaders
2 Win32/FakeXPA 3.3 2 Win32/Small 5.8
Trojans & Droppers
Misc. Potentially Trojan Downloaders
3 Win32/VBInject 2.3 3 Win32/Delf 5.1
Unwanted Software & Droppers
Miscellaneous Miscellaneous
4 Win32/Winwebsec 2.0 4 Win32/Startpage 4.2
Trojans Trojans
Misc. Potentially Misc. Potentially
5 Win32/Obfuscator 1.9 5 Win32/Obfuscator 3.2
Unwanted Software Unwanted Software
Trojan Downloaders
6 Win32/Pdfjsc Exploits 1.4 6 Win32/Banload 2.8
& Droppers
Trojan Downloaders Password Stealers
7 Win32/Small 1.3 7 Win32/Bancos 2.0
& Droppers & Monitoring Tools
Password Stealers & Miscellaneous
8 Win32/Bancos 1.3 8 Win32/Agent 1.1
Monitoring Tools Trojans
Miscellaneous Trojan Downloaders
9 Win32/Swif 1.2 9 Win32/Microjoin 1.1
Trojans & Droppers
Misc. Potentially Trojan Downloaders
10 WinNT/Citeary 1.1 10 Win32/Ciucio 1.0
Unwanted Software & Droppers

 Overall, sites that hosted the top 10 families constituted 76.9 percent of
all malware impressions in the first half of the year and 71.6 percent in
the second half.

 Miscellaneous Potentially Unwanted Software consistently accounts for

between two-thirds and three-fourths of all malware impressions in most
periods, primarily because of Win32/MoneyTree. MoneyTree has been
the malware family responsible for the largest number of malware
impressions during every six-month period since 1H09.

 Document exploit downloads blocked by the SmartScreen filter

decreased from 1.9 percent of the total in 1H10 to 0.96 percent in 2H10.
This decrease correlates with the decline in document exploit detections
in favor of Java exploits, as shown in Figure 6 on page 19.

 Win32/VBInject, Win32/Obfuscator, Win32/Pdfjsc, Win32/Small,

Win32/Startpage, and Win32/Swif are all generic detections for

64
 collections of unrelated threats that share certain identifiable
characteristics.

Global Distribution of Malware Hosting Sites

Figure 41 shows the geographic distribution of malware hosting sites reported to
Microsoft in 2010.

Figure 41. Malware distribution sites per 1,000 Internet hosts for locations around the world in 1H10 (top) and 2H10
(bottom)

65
  As with phishing sites, the worldwide distribution of sites that host
malware remained largely consistent between periods.

Drive-By Download Sites

A drive-by download site is a website that hosts one or more exploits that target
vulnerabilities in web browsers and browser add-ons. Users with vulnerable
computers can be infected with malware simply by visiting such a website, even
without attempting to download anything.

Search engines such as Microsoft Bing™ have taken a number of measures to

help protect users from drive-by downloads. Bing analyzes websites for exploits
as they are indexed and displays warning messages when listings for drive-by
download pages appear in the list of search results. (See Drive-By Download Sites
in the “Reference Guide” section of the Security Intelligence Report website for
more information about how drive-by downloads work and the steps Bing takes
to protect users from them.)

The information in this section was generated from an analysis of the country
code top-level domains (ccTLDs) of the websites in the Bing index that hosted
drive-by download pages in 2010.

66
Figure 42. Percentage of websites in each country-code top-level domain (ccTLD) that hosted drive-by pages in 2Q10
(top) and 4Q10 (bottom)

 In 2H10, drive-by download pages appeared on about 2.4 of every 1,000

search results pages displayed to users during that time.

 Overall, the most heavily infected ccTLDs were small ones. Small TLDs
are susceptible to large swings in infection rates because of their size. For
example, if a major ISP in a small country or region were to become
compromised by an attacker, a large percentage of the domains in the
associated ccTLD could be affected.

67
  Figure 42 does not reflect the physical locations of hosted sites; not all
ccTLD sites are hosted in the locations to which the ccTLDs themselves
are assigned. However, most ccTLD sites are targeted at Internet users in
a particular country/region and are typically written in an appropriate
language, so Figure 42 can be considered a reasonable indicator of how
users in different parts of the world are more or less at risk of
encountering drive-by download pages.

68
Appendix A: Threat Naming
Conventions

The MMPC malware naming standard is derived from the Computer Antivirus
Research Organization (CARO) Malware Naming Scheme, originally published in
1991 and revised in 2002. Most security vendors use naming conventions that
are based on the CARO scheme, with minor variations, although family and
variant names for the same threat can differ between vendors.

A threat name can contain some or all of the components seen in Figure 43.
Figure 43. The Microsoft malware naming convention

The type indicates the primary function or intent of the threat. The MMPC
assigns each individual threat to one of a few dozen different types based on a
number of factors, including how the threat spreads and what it is designed to
do. To simplify the presentation of this information and make it easier to
understand, the Security Intelligence Report groups these types into 10 categories.
For example, the TrojanDownloader and TrojanDropper types are combined into
a single category, called Trojan Downloaders & Droppers.

The platform indicates the operating environment in which the threat is designed
to run and spread. For most of the threats described in this report, the platform
is listed as “Win32,” for the Win32 API used by 32-bit and 64-bit versions of
Windows desktop and server operating systems. (Not all Win32 threats can run
on every version of Windows, however.) Platforms can include programming
languages and file formats, in addition to operating systems. For example, threats
in the ASX/Wimad family are designed for programs that parse the Advanced
Stream Redirector (ASX) file format, regardless of operating system.

Groups of closely related threats are organized into families, which are given
unique names to distinguish them from others. The family name is usually not

69
 related to anything the malware author has chosen to call the threat. Researchers
use a variety of techniques to name new families, such as excerpting and
modifying strings of alphabetic characters found in the malware file. Security
vendors usually try to adopt the name used by the first vendor to positively
identify a new family, although sometimes different vendors use completely
different names for the same threat, which can happen when two or more
vendors discover a new family independently. The MMPC Encyclopedia
(www.microsoft.com/mmpc) lists the names used by other major security
vendors to identify each threat, when known.

Some malware families include multiple components that perform different tasks
and are assigned different types. For example, the Win32/Frethog family includes
variants designated PWS:Win32/Frethog.C and
TrojanDownloader:Win32/Frethog.C, among others. In the Security Intelligence
Report, the category listed for a particular family is the one that Microsoft security
analysts have determined to be the most significant category for the family
(which, in the case of Frethog, is Password Stealers & Monitoring Tools).

Malware creators often release multiple variants for a family, typically in an effort
to avoid being detected by security software. Variants are designated by letters,
which are assigned in order of discovery—A through Z, then AA through AZ,
then BA through BZ, and so on. A variant designation of “gen” indicates that the
threat was detected by a generic signature for the family rather than as a specific
variant. Any additional characters that appear after the variant provide comments
or additional information.

In the Security Intelligence Report, a threat name consisting of a platform and

family name (for example, “Win32/Taterf”) is a reference to a family. When a
longer threat name is given (for example, “Worm:Win32/Taterf.K!dll”), it is a
reference to a more specific signature or to an individual variant. To make the
report easier to read, family and variant names have occasionally been
abbreviated in contexts where confusion is unlikely. Thus, Win32/Taterf is
referred to simply as “Taterf” on subsequent mention in some places, and
Worm:Win32/Taterf.K simply as “Taterf.K.”

70
Appendix B: Data Sources

Microsoft Products and Services

Data included in the Microsoft Security Intelligence Report is gathered from a wide
range of Microsoft products and services. The scale and scope of this telemetry
allows the Security Intelligence Report to deliver the most comprehensive and
detailed perspective on the threat landscape available in the software industry:

 Bing, the search and decision engine from Microsoft, contains technology
that performs billions of webpage scans per year to seek out malicious
content. Once detected, Bing displays warnings to users about the
malicious content to help prevent infection.

 Windows Live® Hotmail® has hundreds of millions of active email users

in more than 30 countries/regions around the world.

 Forefront Online Protection for Exchange (FOPE) protects the networks

of thousands of enterprise customers worldwide by helping to prevent
malware from spreading through email. FOPE scans billions of email
messages every year to identify and block spam and malware.

 Windows Defender is a program that is available at no cost to licensed

users of Windows that provides real-time protection against pop-ups,
slow performance, and security threats caused by spyware and other
potentially unwanted software. Windows Defender runs on more than
100 million computers worldwide.

 The Malicious Software Removal Tool (MSRT) is a free tool that

Microsoft designed to help identify and remove prevalent malware
families from customer computers. MSRT is primarily released as an
important update through Windows Update, Microsoft Update, and
Automatic Updates. A version of the tool is also available from the
Microsoft Download Center. MSRT was downloaded and executed 3.2
billion times in 1H10, or nearly 600 million times each month on
average. MSRT is not a replacement for an up-to-date antivirus solution

71
 because of its lack of real-time protection and because it uses only the
portion of the Microsoft antivirus signature database that enables it to
target specifically selected, prevalent malicious software.

 Microsoft Forefront Endpoint Protection (formerly Forefront Client

Security) is a unified product that provides protection from malware and
potentially unwanted software for enterprise desktops, laptops, and
server operating systems. Like Windows Live OneCare, it uses the
Microsoft Malware Protection Engine and the Microsoft antivirus
signature database to provide real-time, scheduled, and on-demand
protection.

 Microsoft Security Essentials is a real-time protection product that

combines an antivirus and antispyware scanner with phishing and
firewall protection.

 The Windows Live OneCare safety scanner (http://safety.live.com) is a

free online tool that uses the same definition database as the Microsoft
desktop anti-malware products to detect and remove malware and
potentially unwanted software. The Windows Live OneCare safety
scanner is not a replacement for an up-to-date antivirus solution, because
it does not offer real-time protection and cannot prevent a user’s
computer from becoming infected.

 Microsoft Security Essentials is a basic, consumer-oriented anti-malware

product, offered at no charge to licensed users of Windows, which
provides real-time protection against viruses, spyware, and other harmful
software.

 The SmartScreen filter in Internet Explorer 8 and 9 offers Internet

Explorer users protection against phishing sites and sites that host
malware. Microsoft maintains a database of phishing and malware sites
reported by users of Internet Explorer and other Microsoft products and
services. When a user attempts to visit a site in the database with the
filter enabled, Internet Explorer displays a warning and blocks navigation
to the page.

72
Appendix C: Worldwide
Infection Rates

“Global Infection Rates,” on page 27, explains how threat patterns differ
significantly in different parts of the world. Figure 44 shows the infection rates in
locations with at least 100,000 quarterly MSRT executions in 2010. (CCM is the
number of computers cleaned for every 1,000 executions of MSRT. See the
Security Intelligence Report website for more information about the CCM metric.)
Figure 44. Infection rates (CCM) for locations around the world in 2010, by quarter

Country/Region 1Q10 2Q10 3Q10 4Q10

Albania 5.6 3.3 3.1 2.6
Algeria 2.2 2.1 2.4 2.7
Angola 5.4 4.6 5.5 3.9
Argentina 9.3 9.7 11.4 9.2
Armenia — — — 2.4
Australia 7.2 5.9 6.2 5.5
Austria 3.8 3.0 3.5 3.3
Azerbaijan 2.9 2.6 4.2 2.8
Bahamas, The 7.3 7.2 6.9 5.4
Bahrain 14.9 15.6 13.6 9.0
Bangladesh 1.2 1.5 1.5 1.5
Barbados 2.6 2.2 2.3 1.5
Belarus 2.0 1.3 1.5 1.5
Belgium 10.1 7.0 7.5 6.1
Bolivia 7.7 7.8 7.1 5.7
Bosnia and Herzegovina 13.7 10.7 10.6 8.3
Brazil 26.1 25.8 26.3 20.8
Brunei 7.7 7.0 8.0 6.6
Bulgaria 10.0 9.0 10.1 9.9
Cambodia — — — 1.5

73
 Country/Region 1Q10 2Q10 3Q10 4Q10
Cameroon 3.9 3.2 3.3 2.8
Canada 5.2 4.5 4.9 4.2
Chile 12.9 12.9 14.9 12.5
China 8.1 5.5 4.5 2.9
Colombia 16.0 13.5 12.6 10.0
Costa Rica 16.4 12.6 11.9 13.2
Côte d’Ivoire 3.9 2.3 2.4 1.8
Croatia 20.4 15.8 14.1 13.4
Cyprus 9.9 9.3 9.0 7.9
Czech Republic 7.1 5.5 6.2 8.0
Denmark 6.0 4.1 4.9 3.9
Dominican Republic 8.9 7.4 7.9 6.9
Ecuador 17.3 12.9 12.0 8.9
Egypt 9.7 9.0 10.0 11.4
El Salvador 20.6 20.5 19.1 15.2
Estonia 11.9 6.0 8.1 5.9
Ethiopia — — 1.3 1.0
Finland 3.7 2.1 3.8 2.3
France 15.5 12.4 12.8 9.8
Georgia 7.9 7.1 7.7 7.3
Germany 5.5 4.6 5.6 5.3
Ghana 2.9 1.6 1.5 1.2
Greece 18.7 15.4 17.5 14.0
Guadeloupe 3.5 3.0 3.6 2.8
Guatemala 16.1 13.3 13.2 10.2
Honduras 14.8 12.6 13.9 11.0
Hong Kong S.A.R. 9.4 9.1 8.8 6.3
Hungary 19.4 15.2 14.9 11.1
Iceland 12.5 7.7 7.1 5.9
India 4.6 3.4 4.1 3.2
Indonesia 3.2 2.7 10.8 7.1
Iraq 7.2 6.7 9.8 10.0
Ireland 7.6 6.4 7.3 6.2
Israel 15.2 12.2 13.6 11.0

74
Country/Region 1Q10 2Q10 3Q10 4Q10
Italy 12.0 9.7 10.3 8.9
Jamaica 5.4 3.7 3.6 2.5
Japan 5.1 4.4 4.6 3.3
Jordan 8.6 7.4 8.4 8.7
Kazakhstan 2.5 2.2 2.5 2.8
Kenya 3.4 2.7 2.9 2.5
Korea 34.4 34.4 30.1 40.3
Kuwait 13.2 11.5 14.6 12.0
Latvia 12.4 10.8 10.8 9.4
Lebanon 6.5 5.6 6.0 4.8
Libya 4.4 4.1 4.7 4.4
Lithuania 13.4 10.1 11.2 10.5
Luxembourg 8.2 7.1 7.9 6.9
Macao S.A.R. 3.2 2.8 2.7 2.1
Macedonia, F.Y.R.O. 9.6 8.0 7.7 6.6
Malaysia 7.6 6.2 6.8 5.1
Malta 6.3 5.9 5.8 4.3
Martinique 3.9 3.7 5.0 3.7
Mauritius 4.7 4.8 5.0 4.9
Mexico 23.9 21.4 21.1 17.4
Moldova 3.3 2.0 2.1 1.6
Mongolia 1.7 1.1 1.3 1.0
Montenegro 7.7 5.3 5.7 4.6
Morocco 2.7 1.9 1.9 1.6
Mozambique — — 8.4 6.9
Nepal 2.3 2.0 2.0 1.8
Netherlands 9.0 6.1 7.3 5.8
Netherlands Antilles 3.0 2.5 2.5 2.2
New Zealand 6.6 4.9 5.7 4.9
Nicaragua 13.5 13.8 11.7 9.1
Nigeria 3.5 3.2 3.7 2.8
Norway 6.6 4.7 5.0 3.8
Oman 13.2 10.0 10.3 9.0
Pakistan 2.4 2.1 2.1 1.8

75
 Country/Region 1Q10 2Q10 3Q10 4Q10
Palestinian Authority 5.1 4.5 5.0 4.8
Panama 14.5 11.5 13.6 11.7
Paraguay 4.9 5.1 4.9 3.4
Peru 16.2 19.2 16.7 13.5
Philippines 3.0 3.3 3.5 2.8
Poland 23.6 21.8 22.6 17.3
Portugal 23.0 18.1 19.3 15.6
Puerto Rico 5.0 4.0 4.4 3.6
Qatar 8.9 7.9 7.6 6.4
Réunion 2.8 2.7 4.0 3.0
Romania 6.8 5.7 7.0 5.4
Russia 12.4 11.5 11.1 10.1
Saudi Arabia 17.3 16.8 17.9 15.8
Senegal 3.4 2.6 2.4 1.9
Serbia 7.7 5.3 5.7 4.6
Singapore 9.2 8.0 11.1 11.0
Slovakia 8.8 7.6 8.3 8.5
Slovenia 14.8 10.0 9.8 9.1
South Africa 12.8 11.9 11.8 9.8
Spain 39.2 35.7 36.3 33.2
Sri Lanka 2.3 1.8 2.0 1.7
Sweden 8.0 5.2 5.9 4.4
Switzerland 5.0 4.0 4.7 4.1
Taiwan 29.3 33.5 31.7 24.3
Tanzania 4.3 3.9 4.3 3.1
Thailand 14.6 15.3 17.4 14.5
Trinidad and Tobago 5.6 5.1 6.1 4.6
Tunisia 2.5 1.8 1.9 1.6
Turkey 35.5 36.6 42.4 32.8
Uganda — — 4.4 2.8
Ukraine 4.0 3.6 3.3 3.1
United Arab Emirates 9.5 8.4 9.0 7.5
United Kingdom 7.9 6.7 7.4 8.7
United States 14.8 12.9 13.5 11.6

76
Country/Region 1Q10 2Q10 3Q10 4Q10
Uruguay 3.7 4.4 5.1 3.1
Venezuela 9.9 9.5 9.8 9.7
Vietnam 2.2 2.1 2.1 1.6
Worldwide 10.8 9.6 9.9 8.7

77
 Glossary

adware
A program that displays advertisements. Although some adware can be beneficial
by subsidizing a program or service, other adware programs may display
advertisements without adequate consent.

backdoor trojan
A type of trojan that provides attackers with remote access to infected computers.
Bots are a sub- category of backdoor trojans. Also see botnet.

botnet
A set of computers controlled by a “command-and-control” (C&C) computer to
execute commands as directed. The C&C computer can issue commands directly
(often through Internet Relay Chat [IRC]) or by using a decentralized
mechanism, such as peer-to-peer (P2P) networking. Computers in a botnet are
often called nodes or zombies.

C&C
Short for command and control. See botnet.

CCM
Short for computers cleaned per mille (thousand). The number of computers
cleaned for every 1,000 executions of MSRT. For example, if MSRT has 50,000
executions in a particular location in the first quarter of the year and removes
infections from 200 computers, the CCM for that location in the first quarter of
the year is 4.0 (200 ÷ 50,000 × 1,000).

clean
To remove malware or potentially unwanted software from an infected computer.
A single cleaning can involve multiple disinfections.

command and control

See botnet.

78
definition
A set of signatures that can be used to identify malware by using antivirus or
antispyware products. Other vendors may refer to definitions as DAT files,
pattern files, identity files, or antivirus databases.

disclosure
Revelation of the existence of a vulnerability to a third party.

disinfect
To remove a malware or potentially unwanted software component from a
computer or to restore functionality to an infected program. Compare with clean.

downloader/dropper
See trojan downloader/dropper.

exploit
Malicious code that takes advantage of software vulnerabilities to infect a
computer or perform other harmful actions.

firewall
A program or device that monitors and regulates traffic between two points, such
as a single computer and the network server, or one server to another.

generic
A type of signature that is capable of detecting a variety of malware samples from
a specific family, or of a specific type.

IFrame
Short for inline frame. An IFrame is an HTML document that is embedded in
another HTML document. Because the IFrame loads another webpage, it can be
used by criminals to place malicious HTML content, such as a script that
downloads and installs spyware, into non-malicious HTML pages that are hosted
by trusted websites.

Internet Relay Chat (IRC)

A distributed real-time Internet chat protocol that is designed for group
communication. Many botnets use the IRC protocol for C&C.

keylogger
A program that sends keystrokes or screen shots to an attacker. Also see password
stealer (PWS).

79
 Malicious Software Removal Tool
The Microsoft Windows Malicious Software Removal Tool (MSRT) is designed to
help identify and remove specifically targeted, prevalent malware from customer
computers and is available at no charge to licensed Windows users. The main
release mechanism of MSRT is through Windows Update (WU), Microsoft
Update (MU), or Automatic Updates (AU). A version of the tool is also available
for download from the Microsoft Download Center. MSRT is not a replacement
for an up-to-date antivirus solution, because it specifically targets only a small
subset of malware families that are determined to be especially prevalent. In
addition, MSRT includes no real-time protection and cannot be used to prevent
malware from initially infecting a computer. More details about MSRT are
available at www.microsoft.com/security/malwareremove/default.mspx.

malware
Malware is any software that’s been designed specifically to cause damage to a
user’s computer, server, or network. Viruses, worms, trojans, and spyware are all
types of malware.

monitoring tool
Software that monitors activity, usually by capturing keystrokes or screen images.
It may also include network sniffing software. Also see password stealer (PWS).

password stealer (PWS)

Malware that is specifically used to transmit personal information, such as user
names and passwords. A PWS often works in conjunction with a keylogger. Also
see monitoring tool.

payload
The actions conducted by a piece of malware for which it was created. Payloads
can include, but are not limited to, downloading files, changing system settings,
displaying messages, and logging keystrokes.

phishing
A method of credential theft that tricks Internet users into revealing personal or
financial information online. Phishers use phony websites or deceptive email
messages that mimic trusted businesses and brands to steal personally
identifiable information (PII), such as user names, passwords, credit card
numbers, and identification numbers.

80
phishing impression
A single instance of a user attempting to visit a known phishing page with
Internet Explorer 7, 8, or 9, and being blocked by the Phishing Filter or
SmartScreen filter. Also see malware impression.

pop-under
A webpage that opens in a separate window that appears beneath the active
browser window. Pop-under windows are commonly used to display
advertisements.

potentially unwanted software

A program with potentially unwanted functionality that is brought to the user’s
attention for review. This functionality may affect the user’s privacy, security, or
computing experience.

remote control software

A program that provides access to a computer from a remote location. Such
programs are often installed by the computer owner or administrator and are
only a risk if unexpected.

rogue security software

Software that appears to be beneficial from a security perspective but that
provides limited or no security capabilities, generates a significant number of
erroneous or misleading alerts, or attempts to socially engineer the user into
participating in a fraudulent transaction.

rootkit
A program whose main purpose is to perform certain functions that cannot be
easily detected or undone by a system administrator, such as hiding itself or
other malware.

signature
A set of characteristics that can identify a malware family or variant. Signatures
are used by antivirus and antispyware products to determine whether a file is
malicious or not. Also see definition.

social engineering
A technique that defeats security precautions by exploiting human
vulnerabilities. Social engineering scams can be both online (such as receiving
email messages that ask the recipient to click the attachment, which is actually
malware) and offline (such as receiving a phone call from someone posing as a
representative from one’s credit card company). Regardless of the method

81
 selected, the purpose of a social engineering attack remains the same—to get the
targeted user to perform an action of the attacker's choice.

spam
Bulk unsolicited email. Malware authors may use spam to distribute malware,
either by attaching the malware to email messages or by sending a message
containing a link to the malware. Malware may also harvest email addresses for
spamming from compromised machines or may use compromised machines to
send spam.

spyware
A program that collects information, such as the websites a user visits, without
adequate consent. Installation may be without prominent notice or without the
user’s knowledge.

tool
Software that may have legitimate purposes but may also be used by malware
authors or attackers.

trojan
A generally self-contained program that does not self-replicate but takes
malicious action on the computer.

trojan downloader/dropper
A form of trojan that installs other malicious files to a computer that it has
infected, either by downloading them from a remote computer or by obtaining
them directly from a copy contained in its own code.

virus
Malware that replicates, typically by infecting other files in the computer, to
allow the execution of the malware code and its propagation when those files are
activated.

vulnerability
A weakness, error, or poor coding technique in a program that may allow an
attacker to exploit it for a malicious purpose.

worm
Malware that spreads by spontaneously sending copies of itself through email or
by using other communication mechanisms, such as instant messaging (IM) or
peer-to-peer (P2P) applications.

82
Threat Families Referenced in
This Report

The definitions for the threat families referenced in this report are adapted from
the Microsoft Malware Protection Center encyclopedia
(www.microsoft.com/security/portal), which contains detailed information about
a large number of malware and potentially unwanted software families. See the
encyclopedia for more in-depth information and guidance for the families listed
here and throughout the report.

Win32/Agent. A generic detection for a number of trojans that may perform

different malicious functions. The functionality exhibited by this family is highly
variable.

Win32/Alureon. A data-stealing trojan that gathers confidential information

such as user names, passwords, and credit card data from incoming and outgoing
Internet traffic. It may also download malicious data and modify DNS settings.

Win32/Autorun. A family of worms that spreads by copying itself to the mapped

drives of an infected computer. The mapped drives may include network or
removable drives.

Win32/BaiduSobar. A Chinese-language Web browser toolbar that delivers pop-

up and contextual advertisements, blocks certain other advertisements, and
changes the Internet Explorer search page.

Win32/Bancos. A data-stealing trojan that captures online banking credentials

and relays them to the attacker. Most variants target customers of Brazilian
banks.

Win32/Banload. A family of trojans that download other malware. Banload

usually downloads Win32/Banker, which steals banking credentials and other
sensitive data and sends it back to a remote attacker.

Win32/Ciucio. A family of trojans that connect to certain websites in order to

download arbitrary files.

83
 WinNT/Citeary. A kernel mode driver installed by Win32/Citeary, a worm that
spreads to all available drives including the local drive, installs device drivers and
attempts to download other malware from a predefined website.

Win32/ClickPotato. A program that displays popup and notification-style

advertisements based on the user’s browsing habits.

Win32/Conficker. A worm that spreads by exploiting a vulnerability addressed

by Security Bulletin MS08-067. Some variants also spread via removable drives
and by exploiting weak passwords. It disables several important system services
and security products, and downloads arbitrary files.

Win32/Cutwail. A trojan that downloads and executes arbitrary files, usually to

send spam. Win32/Cutwail has also been observed to download the attacker tool
Win32/Newacc.

JS/CVE-2010-0806. A detection for malicious JavaScript that attempts to exploit

the vulnerability addressed by Microsoft Security Bulletin MS10-018.

Win32/Delf. A detection for various threats written in the Delphi programming

language. The behaviors displayed by this malware family are highly variable.

Win32/FakeCog. A rogue security software family distributed under the names

Defense Center, AntiMalware, and many others.

Win32/FakePAV. A rogue security software family that masquerades as

Microsoft Security Essentials.

Win32/FakeRean. A rogue security software family distributed under a variety of

randomly generated names, including Win 7 Internet Security 2010, Vista
Antivirus Pro, XP Guardian, and many others.

Win32/FakeSpypro. A rogue security software family distributed under the

names Antivirus System PRO, Spyware Protect 2009, and others.

Win32/FakeVimes. A rogue security software family distributed under the

names Ultra Antivir 2009, Extra Antivirus, Virus Melt, and many others.

Win32/FakeXPA. A rogue security software family distributed under the names

Antivirus 7, Personal Security, AntiVir2010, Antivirus BEST, Green AV, MaCatte,
and many others.

84
Win32/Frethog. A large family of password-stealing trojans that target
confidential data, such as account information, from massively multiplayer online
games.

Win32/Hamweq. A worm that spreads through removable drives, such as USB

memory sticks. It may contain an IRC-based backdoor that enables the computer
to be controlled remotely by an attacker.

Win32/Hotbar. Adware that displays a dynamic toolbar and targeted pop-up ads
based on its monitoring of Web-browsing activity.

Win32/Keygen. A generic detection for tools that generate product keys for
illegally obtained versions of various software products.

Win32/Microjoin. A generic detection for tools that bundle malware files with
clean files in an effort to deploy malware without being detected by security
software.

Win32/MoneyTree. A family of software that provides the ability to search for

adult content on local disks. It may also install other potentially unwanted
software, such as programs that display pop-up ads.

Win32/Nbar. A program that may display advertisements and redirect user

searches to a certain website. It may also download malicious or unwanted
content into the system without user consent.

Win32/Obfuscator. A generic detection for programs that have had their

purpose disguised to hinder analysis or detection by anti-virus scanners. Such
programs commonly employ a combination of methods, including encryption,
compression, anti-debugging and anti-emulation techniques.

Win32/Onescan. A Korean-language rogue security software family distributed

under the names One Scan, Siren114, EnPrivacy, PC Trouble, My Vaccine, and
others.

Win32/Parite. A family of viruses that infect .exe and .scr executable files on the
local file system and on writeable network shares.

Win32/Pdfjsc. A family of specially crafted PDF files that exploit Adobe Acrobat
and Adobe Reader vulnerabilities. Such files contain malicious JavaScript that
executes when the file is opened.

JS/Pornpop. A generic detection for specially-crafted JavaScript-enabled objects

that attempt to display pop-under advertisements, usually with adult content.

85
 Win32/RealVNC. A management tool that allows a computer to be controlled
remotely. It can be installed for legitimate purposes but can also be installed from
a remote location by an attacker.

Win32/Renos. A family of trojan downloaders that install rogue security

software.

Win32/Rimecud. A family of worms with multiple components that spread via

fixed and removable drives and via instant messaging. It also contains backdoor
functionality that allows unauthorized access to an affected system.

Win32/Rustock. A multi-component family of rootkit-enabled backdoor trojans

that were first developed around 2006 to aid in the distribution of spam email.

Win32/Sality. A family of polymorphic file infectors that target executable files

with the extensions .scr or .exe. They may execute a damaging payload that
deletes files with certain extensions and terminates security-related processes and
services.

JS/ShellCode. A generic detection for JavaScript-enabled objects that contain

exploit code and may exhibit suspicious behavior. Malicious websites and
malformed PDF documents may contain JavaScript that attempts to execute code
without the affected user's consent.

Win32/Small. A generic detection for a variety of threats.

Win32/Sogou. A Chinese-language browser toolbar that may display popup

advertisements and may download and install additional components without
user consent.

Win32/Startpage. A detection for various threats that change the configured

start page of the affected user’s web browser, and may also perform other
malicious actions.

Win32/Swif. A generic detection for maliciously-crafted SWF (Small Web

Format) files. SWF files are commonly used for graphics and video online and
are developed for the Adobe Flash platform.

Win32/Taterf. A family of worms that spread through mapped drives to steal

login and account details for popular online games.

Win32/VBInject. A generic detection for obfuscated malware. The loader is

written in Microsoft Visual Basic® and the malicious code, which may have
virtually any purpose, is encrypted.

86
Win32/Vobfus. A family of worms that spreads via network drives and
removable drives and download/executes arbitrary files. Downloaded files may
include additional malware.

Win32/Winwebsec. A rogue security software family distributed under the

names Winweb Security, System Security, and others.

Win32/Zwangi. A program that runs as a service in the background and

modifies Web browser settings to visit a particular website.

87
 One Microsoft Way
Redmond, WA 98052-6399
microsoft.com/security

88